Integrated circuit technology relies on transistors to formulate vast arrays of functional circuits. The complexity of these circuits require the use of an ever increasing number of linked transistors. As the number of transistors required increases, the surface space on the silicon chip dwindles. It is one objective then, to construct transistors which occupy less surface area on the silicon chip.
Integrated circuit technology uses transistors conjunctively with Boolean algebra to create a myriad of functional digital circuits, also referred to as logic circuits. In a typical arrangement, transistors are combined to switch or alternate an output voltage between just two significant voltage levels, labeled logic 0 and logic 1. Most logic systems use positive logic, in which logic 0 is represented by zero volts, or a low voltage, e.g., below 0.5 V; and logic 1 is represented by a higher voltage.
One method in which these results are achieved involves Complementary Metal-Oxide Semiconductor (CMOS) technology. CMOS technology comprises a combination of oppositely doped Metal-Oxide Semiconductor Field-Effect Transistors (MOSFETs) to achieve the switching mechanism between voltage levels associated with logic 0 and that of logic 1. This configuration is likewise referred to as an inverter. Conventional CMOS inverters consume an appreciable amount of chip surface area, even despite ongoing reductions in the critical dimensions that are achievable with conventional photolithography techniques. The critical dimension (F) represents the minimum lithographic feature size that is imposed by lithographic processes used during fabrication.
MOSFETs are prevalent in integrated circuit technology because they generally demand less power than their counterpart, bipolar junction transistors (BJTs). BJTs, on the other hand, also possess certain advantages over MOS transistors, such as better switching speed. Therefore, it is another objective, and attempts have been made, to combine the technological designs of bipolars and mosfets in an effort to maximize the benefits of both transistor types.
Various types of lateral transistors have been historically described and utilized in complementary metal-oxide semiconductor (CMOS) technology. Lateral bipolar transistors have received renewed interest with the advent of bipolar complementary metal-oxide semiconductor (BiCMOS) technologies. Recently the action of newer devices has been described in new terms and a more careful distinction made between the different types of transistor action possible. Both gate-body connected MOS transistors and gated lateral bipolar transistors have been described. The term gate-body connected transistors is used to describe vertical or other device structures where the body of the MOS transistor also serves as the base of a bipolar transistor but each device functions separately as a normal transistor and MOS transistor action is dominant. Applying the gate voltage to the body serves primarily to change the threshold voltage of the MOS transistor.
Other structures are possible where the gate and base are common and the bipolar transistor and MOS transistor are in parallel but the bipolar transistor current is dominant. In a gated lateral transistor, not only the structures but also the operation is merged and most current flows along a surface under the gate in either MOS or bipolar operation. In the case of a gated lateral bipolar transistor, at low gate voltages around threshold (V.sub.t), they can act as gate-body connected MOS transistors. At higher input voltages, V.sub.t or more, the bipolar action can dominate and they are more appropriately described as gated lateral bipolar transistors.
Much effort has been placed into the study of these merged transistor structures. Both vertical and lateral structures have been studied . These studies do not look to solutions for conserving precious die space in the fabrication of integrated circuits. Likewise, previous efforts have not been able to maximize low power operation and simultaneously maximize switching speeds. It is desirable then to invent structures, circuits and methods which can accommodate the faster switching speed and low power consumption. Any improved configuration of transistor structure should desirably remain fully integrateable with prevalent integrated circuit design.